CN113099261B

CN113099261B - Node processing method and device and node processing system

Info

Publication number: CN113099261B
Application number: CN202110457596.8A
Authority: CN
Inventors: 孙袁袁
Original assignee: Shanghai Bilibili Technology Co Ltd
Current assignee: Shanghai Bilibili Technology Co Ltd
Priority date: 2021-04-27
Filing date: 2021-04-27
Publication date: 2023-04-18
Anticipated expiration: 2041-04-27
Also published as: CN113099261A

Abstract

The application provides a node processing method and a node processing device, wherein the node processing method is applied to a node distribution server and comprises the steps of obtaining a plurality of initial nodes corresponding to the node distribution server and classifying the plurality of initial nodes according to position information; under the condition that the corresponding initial node exists in the position information according to the classification result, determining a plurality of candidate nodes corresponding to each network service party of the position information from the plurality of initial nodes; and determining a target weight value of each candidate node according to a preset dimension, and sending the target weight value of each candidate node and the domain name address corresponding to each candidate node to a node resolution server.

Description

Node processing method and device and node processing system

Technical Field

The application relates to the technical field of computers, in particular to two node processing methods. The present application also relates to two node processing apparatuses, a node processing system, a computing device, and a computer-readable storage medium.

Background

With the development of communication technology, network video live broadcast is particularly popular among people due to the characteristics of intuition, rich content and the like. In the existing live broadcast system, usually, a main broadcast client uploads a live broadcast video stream to a live broadcast service node (i.e., an edge computing node) for receiving a push stream, and the edge computing node forwards the video stream to a CDN (Content Delivery Network) of each home. However, at present, in the actual stream pushing process of the anchor client, some stream pushing is not smooth due to poor quality of the selected edge computing node, which causes the problem that the user is blocked or cannot watch the stream, and greatly affects the watching effect of the user.

Disclosure of Invention

In view of this, the embodiments of the present application provide two node processing methods. The application relates to two node processing devices, a node processing system, a computing device and a computer readable storage medium, which are used for solving the technical problem that a user watches unsmooth or cannot watch due to the quality of an edge computing node in the prior art.

According to a first aspect of the embodiments of the present application, a node processing method is provided, which is applied to a node distribution server, and includes:

acquiring a plurality of initial nodes corresponding to the node distribution server, and classifying the plurality of initial nodes according to position information;

under the condition that the position information is determined to have the corresponding initial node according to the classification result, determining a plurality of candidate nodes corresponding to each network service party of the position information from the plurality of initial nodes;

and determining a target weight value of each candidate node according to a preset dimension, and sending the target weight value of each candidate node and the domain name address corresponding to each candidate node to a node resolution server.

According to a second aspect of the embodiments of the present application, there is provided a node processing method applied to a node resolution server, including:

receiving a node acquisition request of a target object, wherein the node acquisition request carries a domain name address, position information of the target object and a network service party;

resolving the domain name address to obtain a plurality of initial nodes corresponding to the domain name address;

determining a plurality of candidate nodes corresponding to the position information of the target object and a network service party from the plurality of initial nodes, and determining a target weight value of each candidate node;

and determining a target node based on the target weight value, and sending the target node to the target object, wherein the target weight value is received and obtained from a node distribution server.

According to a third aspect of the embodiments of the present application, there is provided a node processing system, including a node allocation server and a node resolution server, wherein,

the node distribution server is configured to acquire a plurality of initial nodes corresponding to the node distribution server and classify the plurality of initial nodes according to position information; under the condition that the position information is determined to have the corresponding initial node according to the classification result, determining a plurality of candidate nodes corresponding to each network service party of the position information from the plurality of initial nodes; determining a target weight value of each candidate node according to a preset dimension, and sending the target weight value of each candidate node and a domain name address corresponding to each candidate node to a node resolution server;

the node distribution server is also configured to receive a domain name address acquisition request sent by a target object, and return the domain name address to the target object based on the acquisition request;

the node analysis server is configured to receive a node acquisition request of a target object, wherein the node acquisition request carries a domain name address, position information of the target object and a network server; analyzing the domain name address to obtain a plurality of initial nodes corresponding to the domain name address; determining a plurality of candidate nodes corresponding to the position information of the target object and a network service party from the plurality of initial nodes, and determining a target weight value of each candidate node; and determining a target node based on the target weight value, and sending the target node to the target object, wherein the target weight value is obtained by receiving from the node distribution server.

According to a fourth aspect of the embodiments of the present application, there is provided a node processing apparatus, applied to a node distribution server, including:

the node classification module is configured to acquire a plurality of initial nodes corresponding to the node distribution server and classify the plurality of initial nodes according to the position information;

a candidate node determination module configured to determine, from the plurality of initial nodes, a plurality of candidate nodes corresponding to each network service provider of the location information in a case where it is determined that the corresponding initial node exists in the location information according to the classification result;

the node analysis server comprises a weight value calculation module and a node analysis server, wherein the weight value calculation module is configured to determine a target weight value of each candidate node according to a preset dimension, and send the target weight value of each candidate node and a domain name address corresponding to each candidate node to the node analysis server.

According to a fifth aspect of the embodiments of the present application, there is provided a node processing apparatus, applied to a node resolution server, including:

the system comprises a request receiving module, a network service part and a service part, wherein the request receiving module is configured to receive a node acquisition request of a target object, and the node acquisition request carries a domain name address, position information of the target object and the network service part;

the resolution module is configured to resolve the domain name address to obtain a plurality of initial nodes corresponding to the domain name address;

a weight value determination module configured to determine a plurality of candidate nodes corresponding to the location information of the target object and the network service provider from the plurality of initial nodes, and determine a target weight value of each candidate node;

and the target node determining module is configured to determine a target node based on the target weight value, and send the target node to the target object, wherein the target weight value is obtained by receiving from a node distribution server.

According to a sixth aspect of embodiments herein, there is provided a computing device comprising a memory, a processor and computer instructions stored on the memory and executable on the processor, the processor implementing the steps of the node processing method when executing the instructions.

According to a seventh aspect of embodiments herein, there is provided a computer readable storage medium storing computer instructions which, when executed by a processor, implement the steps of the node processing method.

The node processing method and device provided by the application are applied to a node distribution server, and comprise the steps of obtaining a plurality of initial nodes corresponding to the node distribution server, and classifying the plurality of initial nodes according to position information; under the condition that the position information is determined to have the corresponding initial node according to the classification result, determining a plurality of candidate nodes corresponding to each network service party of the position information from the plurality of initial nodes; and determining a target weight value of each candidate node according to a preset dimension, and sending the target weight value of each candidate node and the domain name address corresponding to each candidate node to a node resolution server. Specifically, the node processing method calculates a target weight value of each candidate node corresponding to each network service side of each location information according to a preset dimension, and in subsequent practical application, it can be determined based on the target weight value that a candidate node with better plug flow quality is recommended to the anchor, so that the anchor can improve the plug flow quality based on the candidate node with better plug flow quality, and the technical problem that a user is blocked or cannot watch due to the quality of an edge computing node is solved.

Drawings

FIG. 1 is a schematic diagram of an HTTP DNS scheduling logic of an edge computing node according to an embodiment of the present application;

fig. 2 is a schematic diagram of a DNS scheduling logic of an edge computing node according to an embodiment of the present application;

fig. 3 is a flowchart of a first node processing method according to an embodiment of the present application;

fig. 4 is a schematic diagram of a processing procedure of performing a push flow and a pull flow based on an edge computing node in a node processing method according to an embodiment of the present application;

fig. 5 is a flowchart of a second node processing method according to an embodiment of the present application;

fig. 6 is a flowchart of a third node processing method according to an embodiment of the present application;

fig. 7 is a schematic structural diagram of a first node processing apparatus according to an embodiment of the present application;

fig. 8 is a schematic structural diagram of a second node processing apparatus according to an embodiment of the present application;

fig. 9 is a block diagram of a computing device according to an embodiment of the present application.

Detailed Description

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. This application is capable of implementation in many different ways than those herein set forth and of similar import by those skilled in the art without departing from the spirit of this application and is therefore not limited to the specific implementations disclosed below.

The terminology used in the one or more embodiments of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the one or more embodiments of the present application. As used in one or more embodiments of the present application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used in one or more embodiments of the present application refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It will be understood that, although the terms first, second, etc. may be used herein in one or more embodiments of the present application to describe various information, these information should not be limited by these terms. These terms are only used to distinguish one type of information from another. For example, a first can also be referred to as a second and, similarly, a second can also be referred to as a first without departing from the scope of one or more embodiments of the present application. The word "if" as used herein may be interpreted as "at" ... "or" when ...or" in response to a determination ", depending on the context.

First, the noun terms to which one or more embodiments of the present application relate are explained.

Live streaming: live audiovisual data transmission that can be transmitted as a steady and continuous stream over a network for viewing by an audience.

Live broadcasting and stream pulling: the method is a process of pulling a live stream to a source station designated by a user through a live broadcast cloud platform.

DNS (domain name resolution service): i.e. services that convert domain names to IP. Com, for example, the IP of the actual access server is found to be 15.1.1.1 through DNS, and then the actual access is only true; mapping relation 1 of domain name and IP to N; a domain name may correspond to multiple server IPs.

HTTPDNS: traffic scheduling solutions based on HTTP protocol and domain name resolution tailored for mobile clients.

CDN (content delivery network): bottlenecks and links which possibly influence the data transmission speed and stability on the Internet are avoided, and the content transmission is faster and more stable.

CDN service provider: a service provider providing a content distribution network.

An edge computing node: a service node for receiving the push flow.

SRT protocol: UDP based transport protocols.

RTMP protocol: real-time message transfer protocol based on TCP.

TCP (Transmission Control Protocol): a connection-oriented, reliable transport layer communication protocol based on byte streams enables bi-directional data transmission after a TCP link is established.

In the existing live broadcast system, generally, a main broadcast pushes a stream to a live edge computing node, the edge computing node forwards the video stream to each CDN, and the quality of the edge computing node receiving the pushed stream directly affects the viewing of a user.

Currently, there are two ways for a user to select an edge computing node for plug flow: DNS scheduling and HTTP DNS scheduling. In practical application, the difference between node scheduling by using DNS and node scheduling by using HTTP DNS is that:

real-time performance: HTTP DNS needs to request the server to obtain real-time available nodes each time it is scheduled, but for DNS scheduling: the DNS registration IP mapping is done before the dispatch, which determines that the node is not updated in real time. The reason is that: both the DNS resolution and the DNS resolution cache of the user side have TTL time, that is, the cache time, which also determines that the DNS resolution and the DNS resolution cache do not have real-time performance. The success rate of DNS resolution may be affected by the failure of the user's home network or the operator's network. So in general, the DNS update frequency will be much lower than the HTTP DNS schedule, looking more at the quality of coverage over a period of time than for a short period of time.

The accuracy of node scheduling: for the HTTP DNS, requests can all be given to different nodes, and the proportions and protocols for controlling node scheduling are completely performed by the central service; for DNS, it is impossible to control accurately, because a domain name can correspond to multiple IP resolutions, which IP is used by a user is determined by the weight of the IP in the DNS resolution + the DNS resolution service provider, and the network condition of the user itself is also related, so the coverage weight of a single node IP in a specific area can only be configured globally, and it is impossible to control accurately.

The embodiments of the present disclosure are only described in detail in the case of DNS scheduling.

In the deployment of the existing push flow node (i.e. edge computing node), generally deployed on a physical machine or container, after the deployment is completed, the operation and maintenance manually registers the IP in the DNS and in the HTTP DNS schedulable list of the service party. And subsequently, judging whether the node is abnormal or not through node detection, and removing the node from the DNS and the HTTP DNS if the node is abnormal, wherein the node detection can be understood as judging whether the flow pushing node can normally provide flow pushing service or not.

In the existing live broadcast stream pushing system, the scheduling quality of nodes (namely edge computing nodes) is not considered: when a user requests to schedule nodes, only whether nodes cover the area of the user can be seen, but from the point of view of node quality, better nodes are considered to be provided for the user, so that the problems of blocking and watching caused by plug flow are reduced. The node quality is not differentiated from the point of view of the push stream protocol: at different nodes, due to differences in machine configuration and performance, the quality of the protocol capable of receiving the stream pushing is different, and after the stream pushing protocol of the user is determined, a quality node with better protocol coverage is given to the protocol. The stream quality is not differentiated from the protocol that pushed the stream: the video stream is pushed to the edge computing node, the edge computing node needs to push the video stream to each CDN, and if the quality of the pushed stream is poor due to the limitation of the outlet bandwidth and the like, the watching of a user is influenced finally. Therefore, theoretically, the quality of the egress bandwidth should also be considered in the node quality, and in the process of pushing out the video stream, the quality which needs to be pushed out according to the protocol statistics. The coverage quality of the nodes in different provinces is not considered: the same node can cover the provinces nearby, and if the anchor stream is in the covered provinces, the nodes can be used. But the nodes are in different provinces, and coverage quality can be different due to network problems. For example, the two nodes a and B can both cover Guangdong and Guangxi 2 provinces, the Guangdong province part of the node a has better coverage quality, and the Guangxi province part of the node B has better coverage quality, so that the user should select the node a theoretically instead of randomly selecting the node AB in the process of pushing flow in Guangdong.

Based on this, in the present application, two node processing methods are provided. The present application relates to two node processing apparatuses, a node processing system, a computing device, and a computer-readable storage medium, which are described in detail in the following embodiments one by one.

Referring to fig. 1, fig. 1 is a schematic diagram illustrating HTTP DNS scheduling logic of an edge computing node according to an embodiment of the present application.

It should be noted that the HTTP DNS scheduling is actually a 302 scheduling logic, and the logic thereof is to allocate a scheduling address to a user instead of a direct domain name address (which may also be referred to as a push flow address or a push flow domain name in this embodiment), and the user initiates an HTTP request (network request) twice to request a real IP push flow address from the scheduling address.

As shown in fig. 1, fig. 1 includes a main broadcast, a central service (i.e. a service of a live platform, which is mainly responsible for allocating a push stream domain name to a user), an HTTP DSN scheduling service, and an edge computing node, specifically:

the method comprises the steps that a main broadcast requests a stream pushing address from a central service, the central service returns a 302 scheduling address to the main broadcast, for example, HTTP:// 2.2.2/ddomain = www.

Referring to fig. 2, fig. 2 is a schematic diagram illustrating DNS scheduling logic of an edge computing node according to an embodiment of the present application.

It should be noted that the DNS schedule may be directly assigned to a user for a push flow domain name, and the details will be described by taking an assigned domain name as www.

As shown in fig. 2, fig. 2 includes a main broadcasting, a central service (i.e. a service of a live broadcasting platform, which is mainly responsible for allocating a push flow domain name to a user), a DNS (i.e. a domain name resolution service), and an edge computing node, specifically:

the center service reads the available edge computing node IP and registers the IP in the DNS resolution server, namely, the edge computing node IP:8.8.8.8 was added under the name www.

In practical application, an anchor requests a central service for a stream pushing domain name, the central service allocates a www · ω · com stream pushing domain name to the anchor after receiving a stream pushing domain name request, the anchor performs video stream pushing after receiving the stream pushing domain name, and according to a DNS resolution protocol, first sends a domain name resolution request to a DNS resolution server, and the DNS resolution server returns www · com available edge computing node IP to the anchor under the condition of receiving the domain name resolution request, and the anchor starts to perform stream pushing to an edge computing node after receiving a real edge computing node IP, so that the anchor pushes a video stream to a CDN in real time through the edge computing node IP, and then the video stream is distributed to a user by the CDN for the user to view.

The node processing system provided in the embodiment of the present application is implemented based on a DNS scheduling logic, and specifically, taking fig. 2 as an example, the node processing system includes a node allocation server and a node resolution server, where,

the node distribution server is configured to acquire a plurality of initial nodes corresponding to the node distribution server and classify the plurality of initial nodes according to position information; under the condition that the corresponding initial node exists in the position information according to the classification result, determining a plurality of candidate nodes corresponding to each network service party of the position information from the plurality of initial nodes; determining a target weight value of each candidate node according to a preset dimension, and sending the target weight value of each candidate node and a domain name address corresponding to each candidate node to a node resolution server;

the node analysis server is configured to receive a node acquisition request of a target object, wherein the node acquisition request carries a domain name address, position information of the target object and a network server; resolving the domain name address to obtain a plurality of initial nodes corresponding to the domain name address; determining a plurality of candidate nodes corresponding to the position information of the target object and a network service party from the plurality of initial nodes, and determining a target weight value of each candidate node; and determining a target node based on the target weight value, and sending the target node to the target object, wherein the target weight value is obtained by receiving from the node distribution server.

The node distribution server can be understood as a central service in fig. 2, the node resolution server can be understood as a DNS resolution service in fig. 2, the target object can be understood as an anchor in fig. 2, and the node can be understood as an edge computing node in fig. 2.

Specifically, a plurality of initial nodes corresponding to the node distribution server can be understood as a plurality of available edge computing nodes corresponding to the central service; location information may be understood as geographical location information, such as administrative regions, provinces, city of land, etc.; the network service provider may be understood as a network service provider such as mobile, universal, telecommunication, etc., and may also be referred to as an operator.

In practical application, the central service acquires a plurality of corresponding available edge computing nodes and classifies the available edge computing nodes according to the position information; then judging whether each piece of position information has a corresponding edge computing node according to the classification result, and if so, determining a plurality of candidate nodes corresponding to each operator of each piece of position information from a plurality of available edge computing nodes; and finally, calculating a target weight value of each candidate node according to a preset dimension, and sending the target weight value of each candidate node and the domain name address corresponding to each candidate node to a DNS (domain name system) resolution service to realize the registration of the edge calculation node and the domain name address in the DNS resolution service.

After the registration is completed, the central service receives a domain name address acquisition request sent by the anchor, and based on the acquisition request, the central service returns the domain name address registered in the DNS resolution service to the anchor.

After receiving the domain name address, the anchor obtains a request to a DNS resolution service request node, the DNS resolution service receives the node obtaining request sent by the anchor, the domain name address in the node obtaining request is resolved to obtain a plurality of edge computing nodes corresponding to the domain name address, then a plurality of candidate nodes corresponding to an operator of the position information are determined based on the position information carried in the node obtaining request and the operator, a target weight value of each candidate node is obtained, finally a target node is determined based on the target weight value, and the target node is sent to the anchor, so that the anchor pushes flow through the target node; for example, if the anchor is Anhui mobile, then it is necessary that no Anhui connected nodes can be assigned to the anchor.

In the embodiment of the application, the node distribution server calculates the weight value of each available initial node covering the position of the available initial node in real time on the basis of the preset dimensionality, and under the condition that a user requests the plug flow node, the node analysis server can recommend the plug flow node with higher weight value and better quality for the user on the basis of the weight value of each available initial node covering the position of the available initial node, so that the plug flow efficiency is integrally improved.

Referring to fig. 3, fig. 3 is a flowchart illustrating a first node processing method according to an embodiment of the present application, where the node processing method is applied to a node distribution server, and specifically includes the following steps.

Step 302: and acquiring a plurality of initial nodes corresponding to the node distribution server, and classifying the plurality of initial nodes according to the position information.

The plurality of initial nodes corresponding to the node allocation server can be understood as a plurality of available initial nodes corresponding to the node allocation server. In the embodiment of the present application, a node may also be referred to as a plug flow node or an edge computing node.

The location information may be understood as geographical location information, such as administrative areas, provinces, city of the grade, etc.

In the embodiment of the present application, the location information further includes primary location information and secondary location information, for example, the primary location information is administrative area, and the secondary location information is province.

In specific implementation, after the node distribution server acquires the available plurality of initial nodes, the node distribution server classifies the acquired plurality of initial nodes according to the position information.

For example, the obtained multiple initial nodes are: an initial node 1, an initial node 2, an initial node 3, an initial node 4 and an initial node 5; the position information is administrative regions including north China and east China.

If the initial nodes 1 and 2 belong to the north China area and the initial nodes 3, 4 and 5 belong to the east China area, the initial nodes 1 and 2 can be determined to belong to one class and belong to the north China area after the initial nodes are classified according to the position information; the initial node 3, the initial node 4 and the initial node 5 belong to one class and belong to the east China.

In specific implementation, the position information includes first-level position information and second-level position information, for example, the first-level position information is a central area of china, and the second-level position information is information of a plurality of provinces included in east of china, for example, shanghai, jiangsu, zhejiang, anhui, and the like.

Then, classifying the plurality of initial nodes according to the position information may be understood as classifying the plurality of initial nodes according to the primary position information, and then determining the initial node corresponding to each secondary position information based on the classification result. That is, the initial node corresponding to each secondary location information is the initial node corresponding to the primary location information.

As described above, the initial nodes corresponding to shanghai, jiangsu, zhejiang and anhui are initial node 3, initial node 4 and initial node 5.

In addition, in order to ensure the weight updating frequency of the nodes, the DNS can recommend the nodes with the latest node weight value every time the DNS acquires the nodes; the node distribution server can acquire a plurality of initial nodes corresponding to the node distribution server in real time, so that the weight values of the nodes can be updated in real time in the following process, and the target weight values of the nodes registered in the DNS are ensured to be the latest as far as possible.

Step 304: and under the condition that the corresponding initial node exists in the position information according to the classification result, determining a plurality of candidate nodes corresponding to each network service party of the position information from the plurality of initial nodes.

The network service provider may be understood as a network service provider such as mobile, universal, telecommunication, etc., and may also be referred to as an operator. That is, each initial node corresponds to a different network service party, for example, the initial node 1 is a mobile node, the initial node 2 is a connected node, the initial node 3 is a mobile node, the initial node 4 is a telecommunications node, the initial node 5 is a mobile node, and the like.

In practical application, after classifying a plurality of initial nodes according to the location information, an initial node corresponding to each location information, for example, an initial node corresponding to each province, may be obtained.

Then, in the case that there exists a corresponding initial node for a certain location information, a plurality of candidate nodes corresponding to each network service party for the location information may be determined from the plurality of initial nodes.

For example, the location information is an emblem, the corresponding initial nodes are initial node 3, initial node 4, and initial node 5, and the initial node 3 is a mobile node, the initial node 4 is a telecommunications node, and the initial node 5 is a mobile node. Then, it may be determined that the candidate node corresponding to the mobile server of the location information is: the initial node 3 and the initial node 5, the candidate nodes corresponding to the telecommunication service side are: the initial node 4.

In another embodiment of this specification, the classifying the plurality of initial nodes according to the location information includes:

the initial nodes are initially classified according to the primary position information, and the initial nodes corresponding to the secondary position information are determined based on the incidence relation between the secondary position information and the primary position information;

correspondingly, the determining that the position information has the corresponding initial node according to the classification result includes:

and determining that the secondary position information has a corresponding initial node according to the classification result.

Specifically, the first-level location information may be understood as an administrative region, the second-level location information may be understood as a province, after a plurality of available initial nodes are obtained, the initial nodes are divided according to the administrative region, then the nodes are determined as nodes of each province of the corresponding administrative region, and then a target weight value of the node corresponding to each province in the province is calculated.

Step 306: and determining a target weight value of each candidate node according to a preset dimension, and sending the target weight value of each candidate node and the domain name address corresponding to each candidate node to a node resolution server.

The preset dimensionality can be set according to actual requirements, for example, under the condition that a target weight value of each node needs to be obtained accurately, weight value calculation can be carried out from multiple dimensionalities, and under the condition that the target weight value of each node needs to be determined rapidly, weight value calculation can be carried out from one or two dimensionalities.

Specifically, the determining a target weight value of each candidate node according to a preset dimension includes:

determining an initial weight value and a position coverage value of each candidate node corresponding to each network service party of the position information;

and determining a target weight value of each candidate node corresponding to each network service party of the position information according to the initial weight value and the position coverage value.

Each candidate node corresponding to each network provider of each location information has an initialization weight (i.e., an initial weight value), for example, the initialization weight is 0.

In practical application, acquiring an initial weight value of each candidate node corresponding to each network provider of each position information; meanwhile, the location coverage value of each candidate node corresponding to each network service party of each location information, for example, the location coverage value of the initial node 3 under the movement of Anhui, is obtained.

The initial weight value and the position coverage value of the candidate node are then added to obtain a target weight value of the candidate node.

Expressed by the formula, the weight of each candidate node is calculated as: the single node weight = coverage score + original weight, wherein the single node weight is a target weight value of a single candidate node, the coverage score is a position coverage value, and the original weight is an initialization weight value.

In the embodiment of the present specification, the target weight value of each candidate node may be determined only in one dimension of the position coverage value, which is convenient for quickly obtaining the target weight value in subsequent practical applications, and improves the work efficiency.

Specifically, the determining the location coverage value of each candidate node corresponding to each network service provider of the location information includes:

determining a jitter rate and a disconnection rate of each candidate node corresponding to each network service party of the position information in the position information;

and determining a position coverage value of each candidate node corresponding to each network service party of the position information based on the jitter rate and the disconnection rate.

The jitter rate may be understood as a probability that a network of the edge computing node slightly jitters, and the disconnection rate may be understood as a probability that a network of the node analysis server terminal severely jitters or a server is directly restarted, resulting in disconnection of a live stream.

Specifically, the jitter rate of each candidate node is determined according to the position information, the disconnection rate is further determined, and the position coverage value of each candidate node is determined based on the jitter rate and the disconnection rate of each candidate node.

In practical application, a main broadcast pushes a live stream to an edge computing node, if the main broadcast network is normal, a push stream link of the main broadcast should be kept stable all the time, and if the network of the edge computing node slightly shakes, the live stream on the node may shake.

The anchor stream is pushed to the edge compute node, and if the anchor network is normal, the anchor stream connection should remain connected all the time. If the network of the edge computing node is severely jittered or the server is directly restarted, the live broadcast stream can be disconnected; that is, the degree of network jitter determines the number of the live broadcast streams, and if the edge computing node is restarted, all the live broadcast streams are influenced; furthermore, the node determination method provided by the embodiment of the present application may reversely deduce the quality of the edge computing node by determining how many live streams of an edge computing node are disconnected within a time period.

Based on this, after determining the jitter rate and the disconnection rate of the candidate node according to the method, the coverage quality = 1-jitter rate-disconnection rate, and the higher the score of the coverage quality, the better the quality, which indicates that the jitter and disconnection frequency is less; and further determining the position coverage value of each candidate node under the position information according to the jitter rate and the disconnection rate of the candidate node.

In the embodiment of the application, the position coverage value of each candidate node is calculated through the jitter rate and the disconnection rate of the candidate node, whether each candidate node is stable or not is judged based on the position coverage value, the target weight value of each candidate node is calculated through the position coverage dimension, and then the probability that the stable candidate node is selected is improved.

Furthermore, the determining, at each initial node corresponding to each network service provider of the location information, jitter rate and outage rate of the location information includes:

acquiring the number of data jitters, the number of data processes and the number of data disconnections when each candidate node corresponding to each network service party of the position information is processed within a preset time period;

determining the jitter rate of each candidate node according to the data jitter number and the data processing number, and determining the disconnection rate of each candidate node according to the data disconnection number and the data processing number.

The data jitter number can be understood as the number of the number points jittered in a certain period; the data processing quantity can be understood as the total receiving flow quantity on the edge computing node in a certain period; the data disconnection number can be understood as the number of how many live streams of one edge computing node are disconnected in a certain period.

In practical applications, the logic for judging jitter: under the condition that the network is stable, the quantity of network packets transmitted by the edge computing node for receiving user plug flow is fixed, when the network quality is poor, part of the transmission packets can be stored in a user side, and when a large number of receiving requests can be received after the edge computing node network is normal, the data transmission packets can be received instantly, so that the quantity of the network packets transmitted by the receiving user is subjected to up-and-down jitter; take a live stream, a time period of 5 minutes as an example: the number of the points for receiving the transmission packets of the stream pushing network is calculated once in 10s, 30 points exist in 5 minutes, the average receiving network packet number of the 30 points is calculated, and then the number of the points exceeding 30% of the average value is judged, and the exceeding is the jitter. If the jitter point number is greater than the total point number by 30%, the live stream is jittered in the time period, that is, the algorithm for dithering the edge calculation node is as follows: jitter rate = number of streams jittered/total number of received streams on the edge compute node.

In the above example, if the time period is 5 minutes, for example, 5 live streams of an edge computing node are disconnected, and the total received live streams at the edge computing node are 30 live streams, the algorithm of the edge computing node disconnection rate is: active disconnect ratio = number of disconnected streams/total number of received streams on the edge compute node, i.e., disconnect ratio =5/30.

According to the node processing method provided by the embodiment of the application, the position coverage value of each candidate node is calculated according to the jitter rate and the disconnection rate of the candidate node, whether each candidate node is stable or not is judged according to the position coverage value, the target weight value of each candidate node is calculated according to the position coverage dimension, and therefore the probability that the stable candidate node is selected is improved.

In another case, the determining, according to the initial weight value and the location coverage value, a target weight value of each candidate node corresponding to each network service provider of the location information includes:

and determining a target weight value of a target candidate node corresponding to each network service party of the position information and a target weight value of each other candidate node except the target candidate node according to the initial weight value, the position coverage value and a preset weight value.

Specifically, a target weight value of a target candidate node corresponding to each network service provider of the location information and a target weight value of each other candidate node except the target candidate node are determined according to the initial weight value, the location coverage value and a preset weight value, and the specific implementation steps in the specific implementation are as follows:

determining an initial weight value and a location coverage value of each candidate node corresponding to each network service party of the location information;

determining a candidate weight value of each candidate node corresponding to each network service party of the location information according to the initial weight value and the location coverage value;

and determining a target weight value of a target candidate node corresponding to each network service party of the position information and a target weight value of each other candidate node except the target candidate node according to the candidate weight value and a preset weight value.

For a detailed description of the initial weight value and the location coverage value, reference may be made to the above embodiments, which are not described herein again.

Specifically, after the initial weight value and the location weight value of each candidate node are obtained, the candidate weight value of each candidate node may be obtained through calculation based on the initial weight value and the location coverage value.

In practical applications, the preset weight value may be understood as a weight value set for a target candidate node corresponding to the location information according to the location information.

Still take candidate nodes as initial node 3 and initial node 5 as examples, where initial node 3 is a mobile node in Anhui, initial node 5 is a mobile node in Shanghai, and the location information is Anhui.

Then, the preset weight value can be understood as the security badge according to the position information, and is a target candidate node corresponding to the security badge: anhui Mobile node, weight value set by initial node 3.

In specific implementation, after the preset weight value of the target candidate node is determined, the target weight value of the target candidate node and the target weight values of other candidate nodes except the target candidate node may be calculated according to the candidate weight value and the preset weight value.

In the embodiment of the application, the target weight value of each candidate node is comprehensively calculated from the position coverage value of the candidate node under the position information and the two dimensions of local coverage, so that the calculation accuracy of the target weight value of the candidate node can be improved, and the node with better quality can be recommended to a user based on the target weight value subsequently, so that the quality problem of the plug flow node is solved.

Specifically, after determining the target weight value of each candidate node corresponding to each network service provider of the location information according to the initial weight value and the location coverage value, the method further includes:

and determining a target candidate node in the candidate nodes corresponding to each network service party of the position information according to the position information, and distributing a corresponding preset weight value to the target candidate node.

In practical application, when the preset weight value is determined, a target candidate node is determined from a plurality of candidate nodes according to position information, and then the corresponding preset weight value is distributed to the target candidate node.

Following the above example, it may be determined that the initial node 3 is a target candidate node from the initial node 3 and the initial node 5 according to the location information, and then set a preset weight value for the initial node 3, for example, the preset weight value is 10, 20, and the like.

In the embodiment of the application, when the target weight value of the candidate node is calculated, the local coverage dimension is considered, and the priority of the target candidate node is adjusted from the angle of optimal local node quality, so that when node recommendation is subsequently performed, the local node can be recommended to a local user with high probability, and the node plug flow quality is improved.

Then, in a case that a preset weight value of a target candidate node is determined, the determining, according to the initial weight value, the location coverage value, and the preset weight value, a target weight value of the target candidate node corresponding to each network service provider of the location information, and a target weight value of each other candidate node except the target candidate node includes:

determining a target weight value of a target candidate node corresponding to each network service party of the location information according to the initial weight value, the location coverage value and a preset weight value,

and determining a target weight value of each other candidate node except the target candidate node according to the initial weight value and the position coverage value.

In specific implementation, after the target candidate node and the preset weight value of the target candidate node are determined, the candidate weight value (obtained based on the initial weight value and the position coverage value) of the target candidate node is added with the preset weight value to serve as the target weight value of the target candidate node, and the candidate weight values of other candidate nodes except the target candidate node are served as the target weight values of the target candidate node.

If expressed by a formula, the target weight value of the target candidate node = candidate weight value + preset weight value; each other candidate node except the target candidate node target weight value = candidate weight value.

The method is widely understood that after the candidate weight value of each candidate node in a plurality of candidate nodes is calculated, only the preset weight value is added to the candidate weight value of the target candidate node determined in the plurality of candidate nodes to serve as the target weight value of the target candidate node, and the candidate weight values are still used as the target weight values of other candidate nodes, so that the level of the target candidate node covered locally is increased, and non-cross-region recommendation can be achieved during subsequent node recommendation.

In another embodiment of the present specification, the determining, according to the initial weight value, the location coverage value, and a preset weight value, a target weight value of a target candidate node corresponding to each network service provider of the location information, and a target weight value of each other candidate node except the target candidate node includes:

and determining a target weight value of a target candidate node corresponding to each network service party of the position information and a target weight value of each other candidate node except the target candidate node according to the initial weight value, the position coverage value, the preset weight value and a protocol coverage value.

Specifically, a target weight value of a target candidate node corresponding to each network service provider of the location information and a target weight value of each other candidate node except the target candidate node are determined according to the initial weight value, the location coverage value, the preset weight value and the protocol coverage value, and the specific implementation steps in the specific implementation are as follows:

determining an initial weight value, a position coverage value and a protocol coverage value of each candidate node corresponding to each network service party of the position information;

determining an alternative weight value of a target candidate node corresponding to each network service party of the position information and alternative weight values of other candidate nodes except the target candidate node according to the candidate weight values and a preset weight value;

and determining a target weight value of a target candidate node corresponding to each network service party of the position information and a target weight value of each other candidate node except the target candidate node according to the candidate weight values and the protocol coverage value.

Specifically, the protocol coverage value may be understood as coverage quality of each candidate node under different protocols, and the calculation manner of the protocol coverage value is the same as that of the position coverage value, and the only difference is that the position coverage value is coverage quality of each candidate node under different position information, and the protocol coverage value is coverage quality of each candidate node under different protocols.

In practical applications, the candidate weight value of the target candidate node corresponding to each network service party of the location information and the candidate weight values of each other candidate nodes except the target candidate node are determined according to the candidate weight values and the preset weight values, and the calculation method of the target weight value of the target candidate node corresponding to each network service party of the location information and the calculation method of the target weight value of each other candidate node except the target candidate node are the same as those in the above embodiment, and therefore, the description is omitted here.

In specific implementation, after the candidate weight value of each candidate node is determined, the candidate weight value of each candidate node is added to the protocol coverage value of the candidate node, so that the target weight value of each candidate node is obtained.

In the embodiment of the application, the target weight value of each candidate node is calculated from multiple dimensions of coverage quality, local coverage, protocol quality coverage and the like of the candidate node under the position information, so that the accuracy of the target weight value of each candidate node is further ensured, and the probability value of the node recommended to a user as the node with better quality is further improved.

Specifically, the protocol coverage value is obtained by the following steps:

determining a connection push-in protocol coverage value and a connection push-out protocol coverage value under a connection transmission protocol of each candidate node corresponding to each network service party of the position information; and

determining a connectionless push-in protocol coverage value and a connectionless push-out protocol coverage value under a connectionless transport protocol for each candidate node corresponding to each network service party of the location information;

and determining the protocol coverage value of each candidate node corresponding to each network service party of the position information according to the connection push-out protocol coverage value, the connectionless push-in protocol coverage value and the connectionless push-out protocol coverage value.

The connection transport protocol may be understood as an RTMP protocol, i.e. a TCP-based real-time messaging protocol. The connectionless transport protocol is understood to be the SRT protocol, i.e. the UDP-based transport protocol.

The connection push protocol coverage value can be understood as the push coverage quality of a node in a certain time period based on RTMP protocol statistics; the connection push protocol coverage value can be understood as the push coverage quality of a node in a certain time period based on RTMP protocol statistics.

The connectionless push protocol coverage value may be understood as the push coverage quality of a node within a certain time period based on the statistics of the SRT protocol; connectionless push-out protocol coverage values may be understood as the quality of the push-out coverage of a node over a period of time based on the statistics of the SRT protocol.

In practical application, different protocols can be used for pushing a live stream to the edge computing node, and the edge computing node can also use different protocols for pushing the live stream to each CDN, namely different protocols can be selected for pushing.

Specifically, calculating a pushed protocol coverage value of each candidate node under an RTMP protocol, and a pushed protocol coverage value of each candidate node under an SRT protocol; and then calculating the final protocol coverage value of each candidate node through the pushed and pushed protocol coverage value of each candidate node under the RTMP protocol and the pushed and pushed protocol coverage value of each candidate node under the SRT protocol.

In the embodiment of the application, the sum of the connection push-in protocol coverage value, the connection push-out protocol coverage value, the connectionless push-in protocol coverage value and the connectionless push-out protocol coverage value is used as the protocol coverage value of each candidate node, and then the target weight value of each candidate node is calculated based on the protocol coverage value of each candidate node and the alternative weight value, so that concepts of different coverage qualities of the candidate nodes under different protocols are introduced into the target weight value of each candidate node, the protocol quality is increased, the target weight value of the candidate node is adjusted, and the target weight value of each candidate node is more accurate.

Referring to fig. 4, fig. 4 is a schematic diagram illustrating a processing procedure of performing a push flow and a pull flow based on an edge computing node in a node processing method according to an embodiment of the present application.

As shown in fig. 4, the anchor sends a live stream to the edge computing node to push the stream, so as to realize pushing; the edge computing node pushes the live stream to a CDN service node to realize pushing; the watching user pulls a stream from the CDN service node to watch the live stream; it should be noted that whether each edge computing node is stable or not may be described by pushing a protocol coverage value and pushing a protocol coverage value, if the pushing protocol coverage value is poor, the pushing protocol coverage value is also poor, and the quality of a live stream watched by a user from a CDN live stream is also relatively poor; if the push-out protocol coverage value is poor, the quality of the live streaming watched by the user from the CDN is also relatively poor; if the difference between the push-in protocol coverage value and the push-out protocol coverage value is large, the situation that the node is unstable exists is indicated, or the difference between the push-in link and the push-out link is caused due to the existence of an abnormal situation.

In specific implementation, the determining, according to the connection push-in protocol coverage value, the connection push-out protocol coverage value, the connectionless push-in protocol coverage value, and the connectionless push-out protocol coverage value, a protocol coverage value of each candidate node corresponding to each network server of the location information includes:

acquiring a connection protocol coverage difference value of the connection push-in protocol coverage value and the connectionless push-in protocol coverage value according to the connection push-in protocol coverage value and the connectionless push-in protocol coverage value;

acquiring a connectionless protocol coverage difference value of the connection pushout protocol coverage value and the connectionless pushout protocol coverage value according to the connection pushout protocol coverage value and the connectionless pushout protocol coverage value;

and taking the sum of the connection push-out protocol coverage value, the connectionless push-out protocol coverage value, the connection protocol coverage difference value and the connectionless protocol coverage difference value as the protocol coverage value of each candidate node corresponding to each network server of the position information.

The connection protocol coverage difference value can be understood as a ratio of a push-in protocol coverage value to a push-out protocol coverage value of a node in a certain time period based on RTMP protocol statistics; the connectionless protocol coverage difference may be understood as the ratio of the push-to-push protocol coverage values for a node over a period of time based on SRT protocol statistics.

Specifically, for example, the connection push protocol coverage value is X1, the connectionless push protocol coverage value X2, and the connection protocol coverage difference value is X3, the connection push protocol coverage value is Y1, the connectionless push protocol coverage value Y2, and the connectionless protocol coverage difference value is Y3.

Connection protocol coverage difference X3= absolute value (| X2-X1 |/X1-predetermined threshold);

connectionless protocol coverage difference Y3= absolute value (| Y2-Y1 |/Y1-predetermined threshold). The maximum value between the two is taken through the two formulas, and if the coverage difference value of the connectionless protocol is larger than a preset threshold value, the quality difference of the node is too large.

The node determining method provided by the embodiment of the application determines the connection push-in push-out quality of each initial node through the connection transport protocol, determines the connectionless push-in push-out quality of each initial node, and further determines the push-in push-out coverage quality of each initial node, so that whether the initial node is stable or not can be determined according to the push-in push-out coverage quality of each node in the follow-up process, and the push stream quality of a live stream is improved.

In addition, the determining, for each candidate node corresponding to each network service provider of the location information, a connection push-in protocol coverage value and a connection push-out protocol coverage value under a connection transport protocol includes:

determining a push-in jitter rate and a push-in disconnection rate under a connection transmission protocol, and a push-out jitter rate and a push-out disconnection rate under the connection transmission protocol of each candidate node corresponding to each network service party of the position information;

determining a connection push protocol coverage value of each candidate node corresponding to each network service party of the position information under a connection transmission protocol based on the push jitter rate and the push disconnection rate; and

and determining a connection push-out protocol coverage value of each candidate node corresponding to each network service party of the position information under a connection transmission protocol based on the push-out jitter rate and the push-out disconnection rate.

In a specific implementation, the determining, by each candidate node corresponding to each network service provider of the location information, a connectionless push-in protocol coverage value and a connectionless push-out protocol coverage value under a connectionless transport protocol includes:

determining a push-in jitter rate and a push-in disconnection rate under a connectionless transport protocol, and a push-out jitter rate and a push-out disconnection rate under the connectionless transport protocol for each candidate node corresponding to each network service side of the location information;

determining a connectionless push-in protocol coverage value of each candidate node corresponding to each network service party of the location information under a connectionless transport protocol based on the push-in jitter rate and the push-out disconnection rate; and

and determining a connectionless pushout protocol coverage value of each candidate node corresponding to each network service party of the position information under a connectionless transport protocol based on the pushout jitter rate and the pushout disconnection rate.

Specifically, according to the calculation method of the jitter rate and the disconnection rate, a connection push protocol coverage value of each candidate node under the RTMP protocol is calculated based on the push jitter rate and the push disconnection rate of each candidate node under the RTMP protocol; and calculating a connection pushout protocol coverage value of each candidate node under the RTMP protocol based on the pushout jitter rate and the pushout disconnection rate of each candidate node under the RTMP protocol. Similarly, according to the calculation method of the jitter rate and the disconnection rate, a connectionless push-in protocol coverage value under the SRT protocol of each candidate node is calculated based on the push-in jitter rate and the push-in disconnection rate of each candidate node under the SRT protocol; and calculating a connectionless pushout protocol coverage value under the SRT protocol of each candidate node based on the pushout jitter rate and the pushout disconnection rate of each candidate node under the SRT protocol.

Further, the jitter rate and the disconnection rate are determined by:

Specifically, the calculation modes of the push-in jitter rate and the push-out disconnection rate under the connection transmission protocol and the push-in jitter rate and the push-out disconnection rate under the connectionless transmission protocol are the same as the modes of the jitter rate and the disconnection rate introduced in the calculation of the position coverage rate, and are not repeated here, and the only difference is that the jitter rate and the disconnection rate under the connection transmission protocol and the connectionless transmission protocol are different from the jitter rate and the disconnection rate under the position information based on the actual application scenario. Specifically, calculation is performed according to an actual application scene.

In the embodiment of the application, the protocol coverage value of each candidate node is calculated based on the jitter rate and the disconnection rate of each candidate node under each protocol, so that the target weight value of each candidate node can be adjusted according to the protocol coverage value of each candidate node in the following process, the quality of the candidate node is determined from the protocol coverage quality dimension of the candidate node, and the quality of the target node sent to a user in the following process is ensured.

In another embodiment of the present specification, the determining, according to the initial weight value, the location coverage value, the preset weight value, and a protocol coverage value, a target weight value of a target candidate node corresponding to each network service provider of the location information, and a target weight value of each other candidate node except the target candidate node includes:

and determining a target weight value of a target candidate node corresponding to each network service party of the position information and a target weight value of each other candidate node except the target candidate node according to the initial weight value, the position coverage value, the preset weight value, the protocol coverage value and a network transmission value.

Specifically, a target weight value of a target candidate node corresponding to each network service provider of the location information and a target weight value of each other candidate node except the target candidate node are determined according to the initial weight value, the location coverage value, the preset weight value, the protocol coverage value and the network transmission value, and the specific implementation steps in the specific implementation are as follows:

determining alternative weight values of target candidate nodes corresponding to each network service party of the position information and alternative weight values of other candidate nodes except the target candidate nodes according to the candidate weight values and preset weight values;

determining a directional weight value of a target candidate node corresponding to each network service party of the position information and a directional weight value of each other candidate node except the target candidate node according to the candidate weight values and the protocol coverage value;

determining a network transmission value of a corresponding target candidate node of each network server of the location information and a network transmission value of each other candidate node except the target candidate node;

and determining a target weight value of a target candidate node corresponding to each network service party of the position information and a target weight value of each other candidate node except the target candidate node according to the directional weight value and the network transmission value.

For example, the position coverage value, the protocol coverage value, the candidate weight value, and the directional weight value may be calculated in the above embodiments, and details thereof are not repeated herein.

In practical applications, the network transmission value may be understood as a bandwidth. Specifically, the bandwidth bearing capacity of each candidate node is fixed, and when node allocation is performed, the anchor number needs to be reasonably allocated according to the number of push streams in each area, and the anchor can be allocated to the push stream node with the optimal bandwidth in the area.

Therefore, when the target weight value of each candidate node is calculated, the target weight value of each candidate node is calculated from the dimension of the bandwidth, and the situation that a candidate node with a large bandwidth bearing capacity is recommended for a user to carry out plug flow is avoided, so that the plug flow efficiency is low.

In the embodiment of the application, the target weight value of each candidate node is calculated according to the coverage quality, the local coverage, the protocol quality coverage, the bandwidth and other dimensions of the candidate node under the position information, so that the accuracy of the target weight value of each candidate node is further ensured, and the probability value of the node recommended to a user as the node with better quality is further improved.

Specifically, the network transmission value is obtained by the following steps:

acquiring a target candidate node corresponding to each network service party of the position information in a preset time period, and network carrying capacity, code rate, the maximum target object number and a preset threshold of each other candidate node except the target candidate node;

determining a network transmission value of a target candidate node corresponding to each network service party of the location information according to the network carrying capacity, the code rate, the maximum target object number and the preset threshold, and

and determining the network transmission value of each other candidate node except the target candidate node according to the network bearing capacity, the code rate, the maximum target object number, the preset threshold value and the number of the nodes of each other candidate node except the target candidate node.

In practical application, the maximum bearer bandwidth of each candidate node is used as a threshold for calculation, how many anchor broadcasts are performed by one operator in one area (e.g. province) within a preset time period (e.g. late peak 18-00).

For example, in the initial node 3, when there are 1000 anchor broadcasters in the late peak of Anhui movement, and the average stream-pushing code rate is 4M, the bandwidth consumption is 4000M at the same time, and if the maximum bearer bandwidth of a certain node is 2000M, that is, the node can only bear 500 anchor broadcasters at most, another 500 anchor broadcasters need to be allocated to other nodes according to the ratio.

Specifically, in the case that each candidate node considers the bandwidth dimension, the target weight value of each candidate node is adjusted again from the bandwidth dimension.

In the case of expression by a formula, the target weight value = directional weight + maximum number of users capable of carrying bandwidth amount/code rate at late peak 100;

target weight value of each of the other candidate nodes except the target candidate node = directional weight + (1-amount of bandwidth that can be carried/bitrate + maximum number of users late peak)/N100.

The load-bearing bandwidth amount can be understood as a network load-bearing amount, the maximum number of users at late peak can be understood as a maximum target object number, and N can be understood as a node number of other candidate nodes except for the target candidate node.

It is generally understood that it is best to push the candidate nodes of the local coverage to the anchor, but in the case that the bandwidth carrying capacity of the candidate nodes of the local coverage is heavy, the bandwidth carrying capacity of each candidate node is also considered comprehensively, and the candidate nodes with smaller bandwidth carrying capacity are allocated to the user, so as to ensure the quality of the push stream of the user.

In the embodiment of the application, the coverage quality dimension of the nodes, the protocol coverage quality dimension of the plug flow protocol and the push-out protocol, and the bandwidth dimension of the nodes are saved, the target weight value of each candidate node covering a certain area is comprehensively adjusted, the quality of each candidate node is comprehensively judged based on the target weight value, and the DNS node coverage is regularly updated, so that the probability that the quality of a single node affects the occurrence of the main broadcast flow is reduced, the plug flow quality is improved, and the problems that a user watches the blocked state and cannot watch the main broadcast flow due to the quality of the plug flow nodes are reduced. In addition, after the target weight value of each candidate node is calculated, the candidate nodes are covered by registration in the DNS according to the target weight value, that is, the target weight values of the nodes in the DNS are updated. See table 1 for details.

TABLE 1

Type of record	Line type				Record value
						A	Movement of Anhui	0	0	180	*...
A	Movement of Anhui	0	0	180	*...
						A	Movement of Anhui	0	0	180	*...
A	Movement of Anhui	0	0	180	*...
						A	Movement of Anhui	0	0	180	*...
A	Movement of Anhui	0	0	180	*...

Table 1 is an analysis graph of the DNS, and in practical application, after receiving a domain name address carried in a node acquisition request of the anchor, the DNS can analyze the table, and then acquire a node corresponding to location information of the anchor and having a highest target weight value from the table to the anchor.

In another embodiment of the present description, when the operator of the relevant province does not have node coverage, the DNS resolution will be according to the large operator. For example, in the northwest region, the number of nodes is small, and sometimes corresponding provinces and operator node coverage cannot be provided. When the user cannot find the corresponding province + the analysis of the operator in the pushing process, the user can find the province + the operator to the upper layer, namely simply find the operator coverage.

Specifically, after classifying the plurality of initial nodes according to the location information, the method further includes:

under the condition that the secondary position information is determined to have no corresponding initial node according to the classification result, determining a plurality of initial nodes corresponding to each network provider;

determining the network bearing capacity of each initial node, and sequencing all the initial nodes according to the network bearing capacity of each candidate node;

determining a preset number of candidate nodes from the plurality of initial nodes based on the sorting result to be associated with the secondary position information, and sending the candidate nodes and domain name addresses corresponding to the candidate nodes to the node resolution server.

In specific implementation, under the condition that it is determined that no corresponding initial node exists in provinces (namely, secondary location information), all administrative regions are circularly traversed, the bandwidth threshold of each node of each operator in each administrative region is obtained, and then two nodes with the highest bandwidth thresholds are selected as target nodes and registered in mobile DNS analysis. See table 2 for details.

TABLE 2

Type of record	Line type				Record value
						A	Telecommunications	0	0	180	*...
A	Telecommunications	0	0	180	*...
						A	Telecommunications	0	0	180	*...
A	Telecommunications	0	0	180	*...
						A	Telecommunications	0	0	180	*...
A	Telecommunications	0	0	180	*...

Table 2 is an analytic graph of a DNS, in practical applications, after receiving a domain name address carried in a node acquisition request of a host, the DNS first acquires a target node to the host through a province + an operator, and if the province does not have a corresponding target node, searches for a target node with better quality to the host according to the operator, for example, the host is a Anhui mobile, at this time, searches for a target node with the best bandwidth quality to the host under the mobile, and specifically, may parse the table according to the operator, and then acquires a node with the highest bandwidth threshold from the table to the host.

In this embodiment of the present specification, when it is determined that the province where the anchor is located does not have node coverage based on the province + the operator, the analysis may be performed according to the operator corresponding to the anchor, so as to allocate a target node with better bandwidth quality in an initial node corresponding to the operator corresponding to the anchor, thereby improving the push stream quality of the anchor.

In another embodiment of this specification, after classifying the plurality of initial nodes according to the location information, the method further includes:

under the condition that it is determined that the secondary position information does not have a corresponding initial node according to the classification result, and each network service side of the secondary position information does not have a corresponding initial node, taking the initial node corresponding to a preset network service side as a candidate node corresponding to the secondary position information;

and sending the candidate node and the domain name address corresponding to the candidate node to the node resolution server.

Specifically, the above embodiments are all based on processing logics of three major operators, but there are some minor operators such as hua county, radio and television, etc., all the time, which cannot be covered by nodes of the three major operators. In the embodiment of the application, a BGP node (i.e., a special line node that can accept any operator) is adopted to register in a default line, that is, a line that is taken by default when no operator can be found.

Specifically, several BGP nodes are selected and registered in a default line, covering minor operations, see table 3 specifically.

TABLE 3

Type of record	Line type				Record value
						A	By default	0	0	180	*...
A	By default	0	0	180	*...
						A	By default	0	0	180	*...
A	By default	0	0	180	*...
						A	By default	0	0	180	*...
A	By default	0	0	180	*...

Table 3 is an analysis diagram of the DNS, and in practical applications, after receiving a domain name address carried in a node acquisition request of a host by the DNS and determining a province + an operator of the host, and the operator does not have a corresponding node, the DNS may analyze the table, and then acquire a default node from the table to the host, so as to ensure that the host can perform normal stream pushing and improve user experience.

In another embodiment of the present specification, the method further comprises:

receiving a domain name address acquisition request sent by a target object, and returning the domain name address to the target object based on the acquisition request.

Specifically, the node assignment service end may calculate the target weight value of the available initial node in real time or in a timed manner, which is not limited in this embodiment of the present application.

And the node distribution server calculates the target weight value of the available initial node, and has no substantial association relation with the received domain name address acquisition request sent by the target object.

In practical application, after receiving a domain name address acquisition request sent by a target object, a node allocation server can send a domain name address registered at a node resolution server to the target object, so that a subsequent node resolution server can determine a target node based on the domain name address carried in the node acquisition request sent by the target object.

Referring to fig. 5, fig. 5 shows a flowchart of a second node processing method provided in an embodiment of the present application, which is applied to a node resolution server, and specifically includes the following steps:

step 502: receiving a node acquisition request of a target object, wherein the node acquisition request carries a domain name address and position information of the target object.

Step 504: and analyzing the domain name address to obtain a plurality of initial nodes corresponding to the domain name address.

Step 506: and determining a plurality of candidate nodes corresponding to each network service party of the secondary position information and a target weight value of each candidate node from the plurality of initial nodes based on the position information.

Step 508: and determining a target node based on the target weight value, and sending the target node to the target object.

And the target weight value is received and obtained from a node distribution server.

Specifically, the node acquires the domain name address carried in the request, that is, the domain name address sent to the target object by the node allocation server in the above embodiment may be understood, and the domain name address is registered at the node resolution server.

In practical application, after receiving a node acquisition request of a target object carrying a domain name address and position information of the target object, a node analysis server firstly analyzes the domain name address to obtain a plurality of initial nodes corresponding to the domain name address, then determines a plurality of candidate nodes corresponding to the position information based on the position information, and simultaneously determines a target weight value of each candidate node, so that the candidate node with the highest target weight value is sent to the target object as the target node, and the target object can subsequently carry out plug flow based on the target node, thereby improving the quality of the plug flow.

Referring to fig. 6, fig. 6 is a flowchart illustrating a third node processing method according to an embodiment of the present application, where the method is applied to a central service, and specifically includes the following steps.

The central service may be understood as the node distribution server in the above embodiment.

Step 602: DNS overlay.

In practical application, the DNS overlay may be understood as acquiring a plurality of initial nodes corresponding to the central service, and then selecting a target node based on the provinces, operators, and weights of the plurality of initial nodes.

Step 604: province + operator.

Before the DNS performs target node push, the central service calculates a weight of each initial node corresponding to the DNS, for example, in the case that the initial node has a corresponding province and an operator, a plurality of initial nodes may be analyzed from the province + operator dimension.

Step 606: and selecting a fixed province + operator node.

Specifically, the initial node below each operator of each province is selected to calculate the weight value of each initial node of each operator of each province.

Step 608: the weight 0 is initialized.

Specifically, the weight value of each initial node of each operator per province is set to 0, i.e., the original weight of each initial node.

Step 610: and inquiring province coverage quality.

Specifically, the coverage score of each initial node under the operator corresponding to the province corresponding to the initial node is calculated, for example, if the initial node is an Anhui mobile, the coverage score of the Anhui mobile node under the operator in Anhui province is calculated, and the specific calculation process thereof can be referred to the above embodiment.

Step 612: single node weight 1= coverage score + original weight.

Specifically, the weight of each initial node is equal to the calculated coverage score plus its initialization weight.

Step 614: and (6) local coverage.

Specifically, the weight value of the initial node of the local coverage is adjusted, that is, the weight value of the initial node of the local coverage is added with the fixed score. Following the above example, e.g. locally Anhui, a fixed score is added to Anhui mobile nodes, e.g. 50 fixed scores.

Step 616: node weight 2= node weight 1+ fixed score.

Specifically, a fixed score is added to the initial node of the local overlay, that is, the weight of the initial node of the local overlay is equal to the weight obtained in step 612 plus the fixed score; while the scores of other non-locally covered initial nodes remain the weights obtained in step 612.

Step 618: the protocol coverage quality is queried.

Specifically, the coverage quality of each initial node under the RTMP and SRT protocols is calculated, and the specific calculation method may refer to the above embodiment, which is not described herein again.

Step 620: node weight 3= node weight 2+ X1+ X2+ X3+ Y1+ Y2+ Y3.

Specifically, the weight obtained by calculating the coverage quality of each initial node under the RTMP and SRT protocols is node weight 3, where X1 is the coverage quality under the RTMP protocol pushed, X2 is the coverage quality under the RTMP protocol pushed, X3 is the RTMP protocol pushed coverage quality difference, Y1 is the coverage quality under the SRT protocol pushed, Y2 is the coverage quality under the SRT protocol pushed, and Y3 is the SRT protocol pushed coverage quality difference.

Step 622: and controlling the bandwidth.

Specifically, the target weight value is calculated from the bandwidth dimension of each initial node.

Step 624: local overlay node weight = node weight 3+ maximum number of late peak users able to withstand loan amount/rate + 100.

Specifically, step 624 is to calculate the target weight value of the local coverage node, and the specific calculation manner may refer to the above embodiment.

Step 626: the other node weight = node weight 3+ (1-amount of loan to be tolerated/rate of late peak maximum number of users)/N100.

In particular, the other node weights may be understood as weights of other initial nodes than the local overlay node.

After the target weight value of each initial node under each operator of each province is calculated, the target weight value is registered in a DNS, so that when a subsequent anchor requests a node, the DNS can allocate a high-quality node to the anchor for flow pushing based on the target weight value of each initial node.

Step 628: three operators.

In the practical application process, when the operator of the relevant province does not have the node coverage, the DNS is analyzed according to a large operator. For example, in the northwest region, the number of nodes is small, there are sometimes no corresponding provinces and operator node coverage, and when the corresponding province and operator cannot be found in the actual push flow, the operator coverage can be simply found.

Step 630: and traversing according to administrative regions.

Specifically, all administrative areas are traversed, that is, on a per-operator basis, initial nodes under each operator of each administrative area are traversed, and then two or three target nodes with the highest bandwidth thresholds are selected as the target nodes of the administrative areas to be registered in the DNS. Such as a target mobile node, a target connectivity node and/or a target telecommunications node corresponding to the northwest region.

Step 632: and selecting the maximum bearing bandwidth node of each area.

Specifically, the maximum bearer bandwidth node of each administrative area and the administrative area thereof are selected to be registered in the DNS, for example, the northwest administrative area corresponds to the mobile node 1 and the unicom node 2, and when the subsequent anchor is the northwest administrative area and the mobile operator, the node 1 is pushed to the anchor.

Step 634: the default route.

Specifically, the foregoing are all based on processing logic of an operator, and some special small operators cannot be covered by nodes of three large operators, in this case, corresponding BGP nodes may be set for the small operators to be registered in the DNS, and when an operator corresponding to a anchor is the special operator, the corresponding BGP nodes are pushed to the anchor.

Step 636: and selecting a BGP node for registration.

I.e., registers a particular operator and its corresponding BGP node in the DNS.

In the embodiment of the specification, each initial node corresponding to the central service is registered in the DNS in advance according to the province and/or the operator, and subsequently, in practical application, the DNS can push a suitable node for the central service according to the province where the anchor is located and the operator, so that user experience is improved.

Corresponding to the above method embodiment, the present application further provides an embodiment of a node processing apparatus, and fig. 7 shows a schematic structural diagram of a first node processing apparatus provided in an embodiment of the present application. As shown in fig. 7, the apparatus includes:

a node classification module 702 configured to obtain a plurality of initial nodes corresponding to the node distribution server, and classify the plurality of initial nodes according to the location information;

a candidate node determining module 704 configured to determine, if it is determined from the classification result that there exists a corresponding initial node for the location information, a plurality of candidate nodes corresponding to each network service provider of the location information from the plurality of initial nodes;

the weight value calculation module 706 is configured to determine a target weight value of each candidate node according to a preset dimension, and send the target weight value of each candidate node and a domain name address corresponding to each candidate node to the node resolution server.

Optionally, the weight value calculating module 706 is further configured to:

and determining a target weight value of a target candidate node corresponding to each network service party of the location information and a target weight value of each other candidate node except the target candidate node according to the initial weight value, the location coverage value, the preset weight value and a protocol coverage value.

Optionally, the weight value calculating module 706 is further configured to:

and determining a position coverage value of each initial node corresponding to each network service party of the position information based on the jitter rate and the disconnection rate.

Optionally, the weight value calculating module 706 is further configured to:

determining each candidate node corresponding to each network service party of the position information, and a connection push-in protocol coverage value and a connection push-out protocol coverage value under a connection transmission protocol; and

determining a connectionless push-in protocol coverage value and a connectionless push-out protocol coverage value under a connectionless transport protocol for each candidate node corresponding to each network server of the location information;

Optionally, the weight value calculating module 706 is further configured to:

acquiring a target candidate node corresponding to each network service party of the position information in a preset time period, and network bearing capacity, code rate, the maximum target object number and a preset threshold of each other candidate node except the target candidate node;

and determining the network transmission value of each other candidate node except the target candidate node according to the network bearing capacity, the code rate, the maximum target object number, the preset threshold and the number of the nodes of each other candidate node except the target candidate node.

Optionally, the node classifying module 702 is further configured to:

performing initial classification on the plurality of initial nodes according to the primary position information, and determining a plurality of initial nodes corresponding to the secondary position information based on the initial classification result;

Optionally, the apparatus further comprises:

the first node registration module is configured to determine a plurality of initial nodes corresponding to each network provider under the condition that the secondary position information is determined to have no corresponding initial node according to the classification result;

Optionally, the apparatus further comprises:

the second node registration module is configured to take the initial node corresponding to a preset network service party as a candidate node corresponding to the secondary position information under the condition that the secondary position information does not have a corresponding initial node according to the classification result and each network service party of the secondary position information does not have a corresponding initial node;

Optionally, the apparatus further comprises:

the request receiving module is configured to receive a domain name address acquisition request sent by a target object, and return the domain name address for the target object based on the acquisition request.

The node processing device provided by the embodiment of the application calculates the target weight value of each candidate node corresponding to each network service party of each position information according to the preset dimensionality, and in subsequent practical application, the candidate node with better flow pushing quality can be determined based on the target weight value and recommended to the anchor program, so that the anchor program can improve the flow pushing quality based on the candidate node with better flow pushing quality, and the technical problem that a user is blocked or cannot watch due to the quality of the edge calculation node is solved.

The foregoing is a schematic configuration of the first node processing apparatus of this embodiment. It should be noted that the technical solution of the first node processing apparatus and the technical solution of the first node processing method belong to the same concept, and details of the technical solution of the first node processing apparatus, which are not described in detail, can be referred to the description of the technical solution of the first node processing method.

Corresponding to the above method embodiment, the present application further provides an embodiment of a node processing apparatus, and fig. 8 shows a schematic structural diagram of a second node processing apparatus provided in an embodiment of the present application. As shown in fig. 8, the apparatus includes:

a request receiving module 802, configured to receive a node acquisition request of a target object, where the node acquisition request carries a domain name address, location information of the target object, and a network service provider;

a resolution module 804 configured to resolve the domain name address to obtain a plurality of initial nodes corresponding to the domain name address;

a weight value determining module 806 configured to determine a plurality of candidate nodes corresponding to the location information of the target object and the network service provider from the plurality of initial nodes, and determine a target weight value of each candidate node;

a target node determining module 808 configured to determine a target node based on the target weight value, and send the target node to the target object, where the target weight value is received from a node distribution server.

The node processing device provided by the embodiment of the application comprises a node analysis server, a node analysis server and a node processing server, wherein the node analysis server firstly analyzes a domain name address after receiving a node acquisition request that a target object carries the domain name address and position information of the target object so as to obtain a plurality of initial nodes corresponding to the domain name address, then determines a plurality of candidate nodes corresponding to the position information based on the position information, and simultaneously determines a target weight value of each candidate node, so that the candidate node with the highest target weight value is used as the target node to be sent to the target object, the target object can be subjected to flow pushing based on the target node subsequently, and the flow pushing quality is improved.

The foregoing is a schematic configuration of the second node processing apparatus of this embodiment. It should be noted that the technical solution of the second node processing apparatus and the technical solution of the second node processing method belong to the same concept, and details that are not described in detail in the technical solution of the second node processing apparatus can be referred to the description of the technical solution of the second node processing method.

FIG. 9 illustrates a block diagram of a computing device 900 provided in accordance with one embodiment of the present specification. Components of the computing device 900 include, but are not limited to, a memory 910 and a processor 920. The processor 920 is coupled to the memory 910 via a bus 930, and a database 950 is used to store data.

Computing device 900 also includes access device 940, access device 940 enabling computing device 900 to communicate via one or more networks 960. Examples of such networks include the Public Switched Telephone Network (PSTN), a Local Area Network (LAN), a Wide Area Network (WAN), a Personal Area Network (PAN), or a combination of communication networks such as the internet. Access device 940 may include one or more of any type of network interface (e.g., a Network Interface Card (NIC)) whether wired or wireless, such as an IEEE802.11 Wireless Local Area Network (WLAN) wireless interface, a worldwide interoperability for microwave access (Wi-MAX) interface, an ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a bluetooth interface, a Near Field Communication (NFC) interface, and so forth.

In one embodiment of the present description, the above-described components of computing device 900, as well as other components not shown in FIG. 9, may also be connected to each other, such as by a bus. It should be understood that the block diagram of the computing device structure shown in FIG. 9 is for purposes of example only and is not limiting as to the scope of the description. Those skilled in the art may add or replace other components as desired.

Computing device 900 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., tablet computer, personal digital assistant, laptop computer, notebook computer, netbook, etc.), mobile phone (e.g., smartphone), wearable computing device (e.g., smartwatch, smart glasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or PC. Computing device 900 may also be a mobile or stationary server.

The processor 920 is configured to execute computer-executable instructions, and when the processor 920 executes the computer-executable instructions, the steps of the node processing method are implemented.

The above is an illustrative scheme of a computing device of the present embodiment. It should be noted that the technical solution of the computing device and the technical solution of the node processing method belong to the same concept, and details that are not described in detail in the technical solution of the computing device can be referred to the description of the technical solution of the node processing method.

An embodiment of the present application further provides a computer-readable storage medium, which stores computer instructions, and when the instructions are executed by a processor, the computer-readable storage medium implements the steps of the node processing method described above.

The above is an illustrative scheme of a computer-readable storage medium of the embodiment. It should be noted that the technical solution of the storage medium belongs to the same concept as the technical solution of the above-mentioned node processing method, and details that are not described in detail in the technical solution of the storage medium can be referred to the description of the technical solution of the node processing method.

The foregoing description of specific embodiments of the present application has been presented. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

The computer instructions comprise computer program code which may be in the form of source code, object code, an executable file or some intermediate form, or the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, read-Only Memory (ROM), random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, etc. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.

It should be noted that, for the sake of simplicity, the above-mentioned method embodiments are described as a series of acts or combinations, but those skilled in the art should understand that the present application is not limited by the described order of acts, as some steps may be performed in other orders or simultaneously according to the present application. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required in this application.

In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

The preferred embodiments of the present application disclosed above are intended only to aid in the explanation of the application. Alternative embodiments are not exhaustive and do not limit the invention to the precise embodiments described. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the application and its practical application, to thereby enable others skilled in the art to best understand the application and its practical application. The application is limited only by the claims and their full scope and equivalents.

Claims

1. A node processing method is applied to a node distribution server and comprises the following steps:

acquiring a plurality of initial nodes corresponding to the node distribution server, and classifying the plurality of initial nodes according to position information, wherein the flow pushing quality of the plurality of initial nodes corresponding to the position information which is the same as the position information of a target object is better than the flow pushing quality of the plurality of initial nodes corresponding to the position information which is different from the position information of the target object;

under the condition that the corresponding initial node exists in the position information according to the classification result, determining a plurality of candidate nodes corresponding to each network service party of the position information from the plurality of initial nodes;

determining a target weight value of each candidate node according to a preset dimension, and sending the target weight value of each candidate node and a domain name address corresponding to each candidate node to a node resolution server, wherein the node resolution server determines a target node based on the target weight value and sends the target node to a target object;

the determining a target weight value of each candidate node according to a preset dimension includes:

determining an initial weight value and a location coverage value of each candidate node corresponding to each network service party of the location information; the initial weight value is an initialization weight of each candidate node corresponding to each network service side of the location information, the location coverage value is determined based on a jitter rate and a disconnection rate of each candidate node under different location information, the jitter rate is the number of stream jitters divided by the total number of received streams of the candidate nodes, and the disconnection rate is the number of disconnected streams divided by the total number of received streams of the candidate nodes;

2. The node processing method according to claim 1, wherein the determining a target weight value of each candidate node corresponding to each network service provider of the location information according to the initial weight value and the location coverage value comprises:

and determining a target weight value of a target candidate node corresponding to each network service party of the position information and a target weight value of each other candidate node except the target candidate node according to the initial weight value, the position coverage value and a preset weight value, wherein the preset weight value is a weight value set for the target candidate node corresponding to the position information according to the position information.

3. The node processing method according to claim 2, wherein the determining a target weight value of a target candidate node corresponding to each network service provider of the location information and a target weight value of each other candidate node except the target candidate node according to the initial weight value, the location coverage value and a preset weight value comprises:

determining a target weight value of a target candidate node corresponding to each network service party of the location information and a target weight value of each other candidate node except the target candidate node according to the initial weight value, the location coverage value, the preset weight value and a protocol coverage value, wherein the protocol coverage value is determined based on a connection push-in protocol coverage value, a connection push-out protocol coverage value, a connectionless push-in protocol coverage value and a connectionless push-out protocol coverage value.

4. The node processing method according to claim 3, wherein the determining a target weight value of a target candidate node corresponding to each network service provider of the location information and a target weight value of each other candidate node except the target candidate node according to the initial weight value, the location coverage value, the preset weight value and a protocol coverage value comprises:

determining a target weight value of a target candidate node corresponding to each network service party of the location information and a target weight value of each other candidate node except the target candidate node according to the initial weight value, the location coverage value, the preset weight value, the protocol coverage value and a bandwidth, wherein the bandwidth is determined based on a network carrying capacity, a code rate, a maximum target object number and a preset threshold.

5. The node processing method according to claim 2, wherein the determining a target weight value of a target candidate node corresponding to each network service provider of the location information and a target weight value of each other candidate node except the target candidate node according to the initial weight value, the location coverage value and a preset weight value comprises:

6. The node processing method according to claim 4, wherein the protocol coverage value is obtained by:

and determining the protocol coverage value of each candidate node corresponding to each network service party of the position information according to the sum of the connection push-in protocol coverage value, the connection push-out protocol coverage value, the connectionless push-in protocol coverage value and the connectionless push-out protocol coverage value.

7. The node processing method according to claim 4, wherein the bandwidth is obtained by:

acquiring network bearing capacity, code rate, maximum target object number and a preset threshold of a target candidate node corresponding to each network service party of the position information and each other candidate node except the target candidate node within a preset time period, wherein the network bearing capacity is the maximum bandwidth capacity which can be borne by the target candidate node and each other candidate node except the target candidate node;

determining a bandwidth of a corresponding target candidate node of each network service party of the location information according to the network carrying capacity, the code rate, the maximum target object number and the preset threshold of the corresponding target candidate node of each network service party, and

and determining the bandwidth of each other candidate node except the target candidate node according to the network bearing capacity, the code rate, the maximum target object number, the preset threshold and the number of the nodes of each other candidate node except the target candidate node.

8. The node processing method according to claim 1, wherein said classifying the plurality of initial nodes according to the location information comprises:

correspondingly, determining that the position information has a corresponding initial node according to the classification result comprises:

9. The node processing method according to claim 8, wherein after classifying the plurality of initial nodes according to the location information, further comprising:

determining a plurality of initial nodes corresponding to each network provider under the condition that the secondary position information is determined to have no corresponding initial node according to the classification result;

10. The node processing method according to claim 8, wherein after classifying the plurality of initial nodes according to the location information, further comprising:

11. The node processing method according to any one of claims 1 to 4, further comprising:

12. A node processing method is applied to a node analysis server and comprises the following steps:

analyzing the domain name address to obtain a plurality of initial nodes corresponding to the domain name address;

determining a plurality of candidate nodes corresponding to the position information of the target object and the network service provider from the plurality of initial nodes, and determining a target weight value of each candidate node, wherein the target weight value is determined according to a preset dimension, the preset dimension comprises an initial weight value of each candidate node corresponding to each network service provider of the position information and a position coverage value, the initial weight value is an initialization weight of each candidate node corresponding to each network service provider of the position information, the position coverage value is determined based on a jitter rate and a disconnection rate of each candidate node under different position information, the jitter rate is the number of stream jitters divided by the total number of received streams of the candidate nodes, and the disconnection rate is the number of disconnected streams divided by the total number of received streams of the candidate nodes;

and determining a target node based on the target weight value, and sending the target node to the target object, wherein the target weight value is obtained by receiving from a node distribution server.

13. A node processing system is characterized in that the node processing system comprises a node distribution server and a node analysis server, wherein,

the node distribution server is configured to acquire a plurality of initial nodes corresponding to the node distribution server and classify the plurality of initial nodes according to position information, wherein the push flow quality of the plurality of initial nodes corresponding to the position information which is the same as the position information of the target object is better than the push flow quality of the plurality of initial nodes corresponding to the position information which is different from the position information of the target object; under the condition that the corresponding initial node exists in the position information according to the classification result, determining a plurality of candidate nodes corresponding to each network service party of the position information from the plurality of initial nodes; determining a target weight value of each candidate node according to a preset dimension, and sending the target weight value of each candidate node and a domain name address corresponding to each candidate node to a node resolution server, wherein the node resolution server determines a target node based on the target weight value and sends the target node to a target object; the determining a target weight value of each candidate node according to a preset dimension includes: determining an initial weight value and a position coverage value of each candidate node corresponding to each network service party of the position information; the initial weight value is an initialization weight of each candidate node corresponding to each network service side of the location information, the location coverage value is determined based on a jitter rate and a disconnection rate of each candidate node under different location information, the jitter rate is the number of stream jitters divided by the total number of received streams of the candidate nodes, and the disconnection rate is the number of disconnected streams divided by the total number of received streams of the candidate nodes; determining a target weight value of each candidate node corresponding to each network service party of the position information according to the initial weight value and the position coverage value;

the node analysis server is configured to receive a node acquisition request of a target object, wherein the node acquisition request carries a domain name address, position information of the target object and a network server; analyzing the domain name address to obtain a plurality of initial nodes corresponding to the domain name address; determining a plurality of candidate nodes corresponding to the position information of the target object and a network service party from the plurality of initial nodes, and determining a target weight value of each candidate node, wherein the target weight value is determined according to a preset dimension; and determining a target node based on the target weight value, and sending the target node to the target object, wherein the target weight value is obtained by receiving from the node distribution server.

14. A node processing device, applied to a node distribution server, includes:

a node classification module configured to acquire a plurality of initial nodes corresponding to the node distribution server and classify the plurality of initial nodes according to location information, wherein the push flow quality of the plurality of initial nodes corresponding to the location information that is the same as the location information of a target object is better than the push flow quality of the plurality of initial nodes corresponding to the location information that is different from the location information of the target object;

the node analysis server is used for determining a target node according to the target weight value and sending the target weight value of each candidate node and a domain name address corresponding to each candidate node to a node analysis server; the determining a target weight value of each candidate node according to a preset dimension includes: determining an initial weight value and a location coverage value of each candidate node corresponding to each network service party of the location information; the initial weight value is an initialization weight of each candidate node corresponding to each network service side of the location information, the location coverage value is determined based on a jitter rate and a disconnection rate of each candidate node under different location information, the jitter rate is the number of stream jitters divided by the total number of received streams of the candidate nodes, and the disconnection rate is the number of disconnected streams divided by the total number of received streams of the candidate nodes; and determining a target weight value of each candidate node corresponding to each network service party of the position information according to the initial weight value and the position coverage value.

15. A node processing device, applied to a node resolution server, includes:

the resolution module is configured to resolve the domain name addresses to obtain a plurality of initial nodes corresponding to the domain name addresses;

a weight value determining module configured to determine, from the plurality of initial nodes, a plurality of candidate nodes corresponding to location information of the target object and network service providers, and determine a target weight value of each candidate node, where the target weight value is determined according to a preset dimension, the preset dimension includes an initial weight value of each candidate node corresponding to each network service provider of the location information and a location coverage value, the initial weight value is an initialization weight of each candidate node corresponding to each network service provider of the location information, the location coverage value is determined based on a jitter rate and a disconnection rate of each candidate node under different location information, the jitter rate is a number of stream jitter divided by a total number of received streams of the candidate nodes, and the disconnection rate is a number of disconnected streams divided by a total number of received streams of the candidate nodes;

and the target node determining module is configured to determine a target node based on the target weight value, and send the target node to the target object, wherein the target weight value is received and obtained from a node distribution server.

16. A computing device comprising a memory, a processor, and computer instructions stored on the memory and executable on the processor, wherein the processor implements the steps of the method of any one of claims 1-11 or 12 when executing the computer instructions.

17. A computer-readable storage medium storing computer instructions, which when executed by a processor, perform the steps of the method of any one of claims 1 to 11 or 12.